In some conventional imagers, transparent images are sequentially scanned by projecting light through, or reflecting light from, the image and onto an image sensor, such as a CCD imaging array. The signal from the CCD array is conditioned so that an electronic output is generated from the scanned image. The input image may consist of a typical 35 mm slide or other forms of photographic sources. The CCD array typically consists of pixels disposed in rows and columns and producing an output formed as an array conforming to a standard format, for example, as video signals in the NTSC standard format or the PAL standard format. The array responds to the light from the light that is projected through or reflected from the image source that is focused onto the imager. The resulting electronic signal may be stored for future use or it may be employed to generate a new image.
One problem with prior art methods used to generate images is that the resolution has been limited by the resolution capable from the available image sensors, such as CCD arrays. Typically, a photographic image has several times the resolution of a CCD imager. What is actually on the original image may not be exactly what is seen by the CCD since it only registers image data up to the ability of its resolution. Another problem is that only a small fraction of the area of a CCD array is actually photoactive, i. e. the array is "sparse". To get a more accurate reproduction and thereby greater resolution, the CCD has to "see more".
One method of providing the CCD with the ability to "see more" is disclosed in U.S. Pat. No. 4,638,371, entitled Multiple Exposure of Area Image Sensor Having A Sparse Array of Elements, which teaches that each sensor element is multiply exposed by different pixels of an image. The input image is scanned in such a pattern between element exposures that each digital image pixel has a nearest neighbor digital image pixel that was produced by a different sensor element. By this arrangement, a high quality image can be produced from the digital image even if a sensor becomes defective. A dithered sensor is disclosed that has a plurality of positions so as to create a sub-image. The system disclosed, however, does not disclose a simple, economic, and reliable mechanism for dithering the sensor or the input image. Dithering is the mechanical movement of the sensor or the input image in a predetermined pattern to change the image pixel elements that are sensed by the individual elements in an array of scanning elements.
Solid state image sensors generally have a linear or area organization. An area image sensor offers the advantage of increased integration time for each element. In some applications, a large number of image pixels have to be digitized. For example, to make a high quality color print of a photographic image, something on the order of about two million image pixels must be digitized for each color (red, green and blue) of input image to produce a high quality output signal. With existing technology, typical CCD area image sensors have about 300,000 elements. Thus, each element of an area image sensor must sample a plurality of image pixels. Thus, dithering is a technique to create an image of higher resolution than that which would ordinarily be produced with the sensor's normal capability. Such dithering movement must assure that the image sensor remains precisely located along the optic path so that the image is accurately focused. Thus, it will be appreciated that a simple, economic, and reliable imaging platform support that permits dithering movement of the sensor or the input image must be provided.